Working Papers
Paper 63, November 2012
Moving from
war to peace in the
Zambia–Angola borderlands
Oliver Bakewell
This paper is published by the International Migration Institute (IMI), Oxford Department of International
Development (QEH), University of Oxford, 3 Mansfield Road, Oxford OX1 3TB, UK (www.imi.ox.ac.uk).
IMI does not have an institutional view and does not aim to present one.
The views expressed in this document are those of its independent author.
The IMI Working Papers Series
IMI has been publishing working papers since its foundation in 2006. The series presents
current research in the field of international migration. The papers in this series:
analyse migration as part of broader global change
contribute to new theoretical approaches
advance understanding of the multi-level forces driving migration
Abstract
This paper explores the changing relationship between the people of North-Western Zambia
and the nearby border with Angola, focusing on the period as Angola has moved from war
to peace. Drawing on research conducted between 1996 and 2010, the paper examines how
people’s interactions with the border have changed, focusing on their cross-border
livelihoods, identities and mobility. With the end of the war and the rehabilitation of the
formal border crossing, legal restrictions and practical obstacles to movement have relaxed;
at the same time, the conventions – based on informal, ‘illicit’ understandings between local
officials and inhabitants on both sides of the border – that operated for many years have
been undermined. Hence, there has simultaneously been both an ‘opening’ and ‘closing’ of
the border. Moreover, the breaking of these conventions since the end of the war has
reduced the size of the zone of informal exchange and hybridity, or borderlands.
Non-technical summary
This paper explores the changing relationship between the people of North-Western Zambia
and the nearby border with Angola. It focuses on the time when Angola moved from war to
peace. Drawing on research conducted between 1996 and 2010, the paper examines how
people’s interactions with the border have changed. With the end of the war and the
rehabilitation of the formal border crossing, legal restrictions and practical obstacles to
movement have relaxed. At the same time, the conventions – based on informal, ‘illicit’
understandings between local officials and inhabitants on both sides of the border – that
operated for many years have been undermined. There has been both an ‘opening’ and
‘closing’ of the border. The breaking of these conventions has reduced the size of the zone
of informal exchange, or borderlands.
Keywords: borderlands; Angola; Zambia; refugee policy; national identity; livelihoods;
migration; legal status; socio-legal conventions
Author: Oliver Bakewell, Co-Director, International Migration Institute, University of Oxford;
email: [email protected]
2
IMI Working Papers Series 2012, No. 63
Contents
1 Introduction ............................................................................................................................ 4
2 Conceptual framework ........................................................................................................... 4
3 The study area: the Zambia–Angola borderlands................................................................... 7
4 The changing relations with the border ............................................................................... 11
4.1 Borderland identifications ............................................................................................. 11
4.2 Borderland livelihoods ................................................................................................... 14
4.3 Borderland movements ................................................................................................. 16
5 Conclusion ............................................................................................................................. 17
References ............................................................................................................................... 20
IMI Working Papers Series 2012, No. 63
3
1 Introduction
In 1996, Mwinilunga in the far north-west of Zambia seemed close to the edge of the world,
lying at the gateway of the small pedicle jutting between two of Africa’s large and chaotic
states (Clapham, Herbst, and Mills 2006), DR Congo to the north and east, and Angola to the
west. In particular, Angola was caught in the painful extra-time of its long civil war re-ignited
with the failure of the 1992-3 peace process, most of its Moxico Province that borders
Zambia’s Mwinilunga District was controlled by the UNITA (União Nacional para a
Independência Total de Angola) rebels rather than the Luanda government. Mwinilunga
itself was a reasonably peaceful and quiet district town far from Lusaka and the Zambian
Copperbelt. Much of the excitement came with the tales of smuggling arms and diamonds
across the border and with the adventurers who settled in the area who fed these rumours.
To cross the border to Angola was an adventure that promised only unpredictable outcomes
of encounters and negotiations with UNITA, landmines and all the other dangers of war.
Today, while Mwinilunga remains as far as ever from Lusaka, it is no longer on the
road to nowhere, a dead-end open only to those prepared to risk the chaos beyond. A new
sense of place has been created for Mwinilunga by the conclusive end of the war in Angola
in 2002 with the death of UNITA leader Jonas Savimbi. It now has the potential to connect
Zambia to the riches and opportunities opening up in Angola. Indeed if the Benguela
Railway, which runs from the Atlantic right across to Moxico, is rebuilt, this corner of Zambia
may be transformed. Moreover, instead of it being a frontier zone bordering chaos, now the
Angolan government is controlling the territory right up to the border line, administering
immigration and customs posts, patrolling and exercising some semblance of border
control. Crossing the border has become more routine for those with the right papers,
rather than an adventure into the unknown.
This shift from war to peace in Angola has changed the nature of the border between
Angola and Zambia. In this paper, I explore how this has affected the lives of those residing
in villages in the borderlands, away from the formal crossing points and the centres of
administration of either state. Drawing on data gathered on the Zambian side of the border
over the last 14 years, the paper focuses on the changes in the way people identify
themselves, their livelihoods and their movements between Angola and Zambia. It
distinguishes between the formal policies and practices embedded in law and government
action on the one hand, and the locally negotiated patterns of interaction which determine
the rules of the game in the borderland. By exploring these different levels of borderland
practices, it argues that the end of the war has simultaneously opened and closed the
border for different actors, and in many respects reduced the scale of the zone of transition
and negotiation which marks out the borderland.
2 Conceptual framework
This chapter builds on a growing literature on borderlands drawing from political geography,
political science, and anthropology (for example, see Anderson and O'Dowd 1999; BrunetJailly 2005; Das and Poole 2004; Donnan and Wilson 1999; Kolossov 2005; Zartman 2010).
Baud and Van Schendel (1997: 212) note that earlier studies tended to take a state-centred
view, looking at borders as they were seen from the centre. Instead they call for a ‘view
from the periphery’. As they observe, ‘generally speaking there has always been an
4
IMI Working Papers Series 2012, No. 63
enormous gap between the rhetoric of border maintenance and the daily life in the
borderlands’ (Baud and Van Schendel 1997: 220). In the last decade, this balance has started
to be redressed, particularly within the African continent, with studies by historians,
anthropologists and political scientists exploring how the border shapes the lives of those
within the borderlands – and they in turn help to shape the meaning and extent of the
borderlands (Flynn 1997; Megoran 2006; Miles 2005; Nugent 2002; Raeymaekers 2009;
Zeller 2009).
The borderlands can be described as a ‘zone of transition’ in which ‘cultural, linguistic
and social hybridity emerge’ (Newman 2006: 151), creating a zone in which movement
across the line becomes much easier. Newman writes of this in terms of lessening ‘the shock
of meeting the “other”’. This perhaps does not apply where the border runs through sociocultural and linguistic groups, which is common across the African continent; in such cases,
the ‘other’ is not to be found in the population on the other side. Here, the transition zone
may be more concerned with easing the change in economic and legal regimes as one
moves from one state to another. These are the borders which have often been described
as artificial and irrelevant: a colonial construct with little meaning or legitimacy in the eyes
of ordinary people living near the border. It is this sort of border with which I am concerned
in this paper.
In particular, in those African states where the borderlands are peripheral areas far
from the centres of state power, the formal authority of the state may be stretched very
thinly. Its officials are few and far between and they are in isolated situations with very
limited resources. They can only exercise their authority through a process of negotiation
and accommodation with the local population. Such borderlands may be described as
‘unruly’ (Baud and Van Schendel 1997: 227). Nevertheless, as Raeymaekers argues in his
case study of the ‘unruly’ Congo–Uganda border, this process can result in the emergence of
informal and enduring forms of regulatory authority across the borderland (Raeymaekers
2009: 58). The borderlands are concerned with finding local accommodations and
exchanges which enable the ‘silent encroachment’ of the frontier (Raeymaekers 2009: 57).
While this may be presented as a challenge to the state,
the actions and tactics of the poor do not necessarily undermine state
sovereignty as such, but instead mould and transform it in the process of
generating local livelihoods and places (Raeymaekers 2009: 57).
By tracing out the (transborder) livelihoods of the people living on both sides of the border,
Raeymaekers elaborates the workings of these emergent institutional arrangements, or
‘sub-systems’ of power, in which people can acquire unofficial status to regulate movement,
charge ‘taxes’ and control markets. While such arrangements may acquire some stability
and local legitimacy, he suggests that they fall short of constituting a ‘social contract’.
Instead, they give rise to a hybrid system of regulation that mixes ‘different and often
contradictory legal orders and cultures’ (Raeymaekers 2009: 63). As a result of this complex
system, the outcome of any particular interaction between actors is not the predictable
result of these patterns of power relations but rather ‘specific instances of negotiation and
confrontation’: they are embedded in the moment (Raeymaekers 2009: 63).
In a recent paper, Nugent takes up a similar theme in his analysis of the power
dynamics on the Senegambian and Ghana–Togo borderlands (Nugent 2010a). He also
observes the patterns and regularities in the multifarious interactions across the border,
IMI Working Papers Series 2012, No. 63
5
suggesting that over time these have developed into a complex set of rules of the game
which shape future interactions. To help make sense of this changing landscape and explore
the composition of the ‘rules’ pertaining to any encounter, he distinguishes three elements
which must be taken into account: the social contract, the convention and the moment.
The social contract is concerned with the bargain struck between the state and the
people over whom it claims authority – in particular those who stay within its boundaries. It
is articulated both through the laws and state institutions that provide services and regulate
life within the boundaries of the state. It defines what is seen as legitimate action by both
the state and the people. As Nugent argues, even in the seemingly chaotic environments of
many African states, there is often a ‘surprisingly deep-seated attachment to bureaucratic
rules and behaviour’ (Nugent 2010b: 37). The nature of the social contract will affect the
way that government officials interact with the population in the borderlands.
While the social contract is somewhat abstract and negotiated at a distance from the
borderland, a convention is the locally negotiated and more fluid set of practices which
come to define the acceptable behaviour in the borderlands. For instance, the law may
require one to show identity papers to cross the border, but there may be a convention that
local residents can move freely and they will face no penalties if found on the ‘wrong side’.
In Zambia, the social contract offers no provision by which people who enter the country as
refugees can become citizens. Nevertheless, as I will show below, there are well-established
informal procedures whereby self-settled refugees can acquire national registration papers
that secure the benefits of citizenship (including the right to vote). Such conventions
develop over time through the accumulation of practice in encounters between different
actors in the borderland. This notion seems to capture the ‘institutional creativity and
“bricolage”’ described by Raeymaekers (Raeymaekers 2009: 57) and the local negotiation of
a ‘standard model of interaction’, including a ‘free-trade pact’ on the Benin–Nigeria border
as described by Flynn (1997: 323).
Although such conventions may become almost institutionalised at the local level
(while potentially completely disavowed by the social contract), they are fluid and subject to
continuous renegotiation. In particular at the time of any encounter, while the expectations
and behaviour of the actors involved may be shaped by both the social contract and
convention, the actual outcome will be contingent on the particular circumstances of the
moment. For example, Nugent notes that where the border moves through rural locations
cutting across intensive social and linguistic groups, the social contract is likely to have
reduced weight compared with the conventions worked out by government officials as they
negotiate their presence and function in the local communities (Nugent 2010a: 8). Personal
relationships also come to the fore in shaping encounters. For the conventions to operate,
the protagonists not only need to know the ‘rules of the game’ but also each other, or have
some means of vouching for themselves, for example through the mediation of local people
(Flynn 1997: 322). More generally, it seems reasonable to suggest that the extent to which
the social contract or the existing conventions may take precedence may vary greatly
depending on the location, the particular actors involved, the broader socio-cultural
institutions at play, and the time of the encounter.
The moments of encounters, revealed by observation, interviews and narrative
sources, offer a lens through which the workings of conventions and the social contracts in
the borderlands can be perceived. In what follows, I attempt to apply this framework to one
part of the Zambia–Angola borderlands. I set out to explore how people’s relationship with
6
IMI Working Papers Series 2012, No. 63
the border and the shape of the borderland has changed with this transition from war to
peace.
It is important to note that I am using the term ‘borderlands’ to refer to the spatial
and social zone either side of the line on the ground – moving from the one-dimensional line
to the two-dimensional area. Hence, I take the border to be a political and geographical
‘fact’, marked out on maps and in places inscribed in the ground. My interest is in the social
meaning of that ‘fact’, particularly in the day-to-day lives of those in the borderlands and
how it changes over time. Hence the borderland is a dynamic zone of interaction where the
proximity of the border (the line) brings about distinctive social, political, economic and
cultural exchanges. As for meanings, I am concerned with both the objective meaning of
what difference the presence of the border makes to people’s lives – what they do, how
they can live – and also the subjective meanings concerned with how living close to the
border affects their sense of being – their identity and their possibilities. With this in mind,
in what follows, I focus on three particular transnational aspects of life in the borderlands:
identifications, livelihoods, and movements.
3 The study area: the Zambia–Angola borderlands
The focus of this paper is the northern half of the border with Zambia’s North-Western
Province to the east and Angola’s Moxico Province to the west, often known more generally
as the upper Zambezi. The river Zambezi rises in Mwinilunga District north of the town and
flows west into Angola before turning south and re-entering Zambia at Chavuma in Zambezi
district after which it starts its eastern flow to the Indian Ocean. The majority of the
IMI Working Papers Series 2012, No. 63
7
fieldwork was conducted in Mwinilunga District in a village approximately fifty kilometres
west of the district town and eight kilometres from the border with Angola. This is a remote
area of Zambia about three hundred kilometres from the provincial capital, Solwezi, and
nearly one thousand from the capital, Lusaka.
The people of the area are predominantly Lunda (Ndembu), an ethnic group that
extends across the nearby borders into Angola and the Democratic Republic of Congo
(Pritchett 2001; Turner 1957). The main focus of the study was on a village set within the
territory of the Lunda Senior Chief Kanongesha. This is an area where large numbers of
refugees who fled the war in Angola settled, especially from the early 1980s. The process of
integration and their interest in repatriation were the initial focus of the study (see Bakewell
2000).
The history of the Lunda of Mwinilunga and related peoples is one of migration from
Congo, through Angola and across into Zambia. In broad terms, one can trace a general drift
of population from the west and north into present-day Zambia in response to the slave
trade (both to catch slaves and avoid being caught), and in the hunt for ivory, beeswax and
rubber –and Kanongesha was no exception to this pattern.
Although Portugal considered Angola to be its colony from the fifteenth century, it
controlled very little of the territory until the end of the nineteenth century, and most of
that was on the coastal strip (Henderson 1978:106). The end of the slave trade and the
productivity of the interior encouraged explorers to move east and call for the extension of
Portugal’s direct involvement in Angola to promote trade (Clarence-Smith 1979). However,
the major spur for the expansion of Portugal’s control in Angola was the increasing pressure
from Britain, Belgium and France looking to secure their access to the wealth of raw
materials in the ‘scramble for Africa’. At the Conference of Berlin in 1884/85, Britain and
Portugal agreed the line of the border between their territories except for the upper
Zambezi, including Zambia’s North-Western Province and Angola’s Moxico Province; this
border was only finalised in 1905 (Roberts 1976). The British arrived in the north-west of
Zambia in 1906 when the British South African Company (BSAC) established administrative
posts in Mwinilunga and Zambezi.
At the same time as imposing this international border, the colonialists also mapped
out ethnic boundaries between different groups. In particular the British placed great
emphasis on the role of traditional chiefs (as far as they understood it) and gave them an
important role within the developing colonial state, particularly as intermediaries for
extracting taxes from the population and exerting control. To ensure the chiefs’ cooperation they were given a salary and status within the regime.
The British desire to work through chiefs was initially complicated by chiefdoms
crossing the newly demarcated colonial borders. Although they worked with Kanongesha in
Angola for some time, in the 1920s the British requested that he create a new chieftainship
to cover the Zambian side of the border. This was achieved by installing one of his sons,
Mwilombi, as the first Zambian Kanongesha. The Angolan Kanongesha is still seen as the
senior of the two. Hence, the separation of the Lunda of Zambia from their kin in Angola is a
recent product of the imposition of the border and occurred within the living memory of
some of the older people. A similar story could be told of the Luvale and Mbunda further to
the south (Cheke Cultural Writers 1994; Sangambo and M.K. 1984). The continued contact
8
IMI Working Papers Series 2012, No. 63
between chiefs across the border and the exchange of visits ensures that this history is
known and new generations are aware of where they have come from in the recent past. 1
Despite the ‘artificial’ nature of the international border between Zambia and Angola,
which was clearly the creation of colonialists, from the earliest days the local people have
imbued the line on the ground with meaning and sought to use it to their advantage. The
most obvious example of this was in the avoidance of taxation. One of the key roles of the
British colonial administrator in Zambia was to collect taxes from the villagers, and
frequently people ran across the border to avoid registration. When taxation was first
introduced there was a mass exodus from Zambia into neighbouring Angola and Congo.
Likewise, when the Portuguese introduced their taxes, people moved in the other direction
(Pritchett 2001; von Oppen 1995: 432).
Repression and political violence also played important roles in movement between
the colonies. Across Angola, although slavery was abolished in 1878, it was replaced by a
repressive ‘contract labour’ system. People in Kanongesha recalled the harshness of the
system whereby they were forced to go and work on roads, bridges, railways and other
projects for no wages and even no food.
Every adult had to do forced labour; people would be taken in turn from the
village. A group might work once per week or even for a whole month. No pay
was given but possibly a cup of salt. The workers had to bring their own food. If
people hid in the homes they would be beaten and then forced to work and this
caused many people to come to Zambia (interview 5/2/97).
The policy caused a massive migration of Angolans into neighbouring countries with over
half a million living in exile in 1954 (Bender 1978). The resultant labour shortage increased
the demand for contract labour and the system survived until the start of the Angolan
revolution in 1961 (Henderson 1978).
The development of large towns in the Zambian Copperbelt provided another
incentive for the migration of Angolans. Some came for jobs in the mines but many others
were drawn into the large markets generated by the urban areas. In north-west Zambia the
British administration, cut off from white farmers in the south, ‘was lucky in finding a rural
population, continually swelled by kindred immigrants from neighbouring Portuguese
territory (Angola), which was capable of and most interested in producing substantial
amounts of food for sale’ (von Oppen 1995: 432). Traders moved back and forth from
Angola across Kanongesha to bring meat, fish and other produce to ‘town’ (as the
Copperbelt is known) in exchange for soap, salt, cloth and other manufactured goods.
Similar trade continues today.
The establishment of the border thus introduced a new set of constraints and
opportunities for the people of Kanongesha. It became an important part of peoples’ lives as
they crossed to avoid taxation and violence, to gain protection or to find jobs, markets,
education and health care. The border rapidly became something to be exploited but took
longer to gain any significance as a line between people of different nationalities. During the
colonial era there was very little control of the border according to the accounts of the
villagers who look wistfully back to the time when the British and Portuguese were in charge
1
This ongoing contact was illustrated very clearly in August 2010, when Senior Chief Kanongesha from Zambia
and a number of his councillors went to help with the installation of the new Kanongesha in Angola.
IMI Working Papers Series 2012, No. 63
9
and they did not need any papers to cross. This changed with the independence of Zambia
in 1964 and the war of independence in Angola that started in 1961; these two factors
changed the significance of nationality and the border both for the states and the local
people living in the borderlands.
In 1966 an eastern front was opened in Angola’s war of liberation which stimulated
the first movement of refugees into the west of independent Zambia. Angolan
independence in 1975 presaged the start of the civil war between the MPLA, which took
control of the government from the Portuguese, and UNITA. The war was to last for nearly
thirty years. The vast majority of Angolans who fled to Kanongesha arrived between 1982
and 1986 as UNITA took control of the neighbouring Alto Zambeze district of Angola’s
Moxico Province. Cazombo, the district capital, and the other local towns of Calunda, Lovua
and Cavungu were captured between 1982 and 1986 and many villages were totally
destroyed. Cazombo’s vital link to the rest of Angola, the bridge over the Zambezi, was also
destroyed.
During the most violent phases of the war in Moxico Province in the 1980s, the
numbers of people flowing into Mwinilunga District were at times overwhelming and put
immense strain on the local resources, particularly food. However, throughout the border
area, villagers, officials and other observers frequently commented that the Angolans
rapidly settled in the area, worked hard and caused no problems. Nearly all the villagers
who expressed an opinion said that were very glad to see their relatives come to stay. Even
if the initial difficulties of food were severe, they have long since been forgotten.
The MPLA and UNITA reached a peace agreement in 1991 and presidential elections
were held in 1992 (Pycroft 1994). However, far from putting the seal on the peace, the
elections marked the start of a further decade of war. The results were rejected by the
UNITA leader, Jonas Savimbi, who took his forces back to the bush and launched attacks on
many government-held towns. The extreme violence of this ‘war of the cities’ ebbed from
1994 with the signing of the Lusaka Protocol, which resulted in an uneasy stalemate. This
left the country effectively divided between the patchwork of government-controlled areas
focused on the west and the provincial capitals across the country, and the extensive area
controlled by UNITA, whose strongholds were in the south and east, including that part of
Moxico Province, Alto Zambeze District, that borders Zambia.
It was in 1996, during this period of stalemate, that this research project started. From
1996 to 1998, UNHCR was making plans for the imminent repatriation of refugees from
Zambia. These preparations had to be abandoned as the situation in Angola deteriorated
further and the country returned to full-scale war in late 1999. The government launched a
major offensive in UNITA’s heartland in the south and east of Angola and managed to
capture its administrative centre, Jamba. This displaced tens of thousands more refugees
into Zambia in December 1999 and throughout 2000, including many UNITA functionaries
who had been based in Jamba. With the fall of Jamba, the focus of the conflict shifted to
Moxico Province and it was there that the UNITA leader Jonas Savimbi was killed in February
2002. It was only with his death that the war in Angola came conclusively to an end – a
peace agreement was reached six weeks later (International Crisis Group 2003). The country
held parliamentary elections in 2008 but presidential elections have been repeatedly
delayed and are now planned for 2012.
10
IMI Working Papers Series 2012, No. 63
4 The changing relations with the border
After this brief sketch of the local context for the study, this section explores in more detail
the changing meaning of the border for residents of the borderlands as Angola has moved
from war to peace. It highlights the changing interaction between the social contracts,
conventions and moments drawing on detailed fieldwork largely focused on one
chieftainship on the Zambian side of the border. It looks at how people position themselves
in relation to Angola and Zambia, through the use of papers and expressions of national
identity. It then shows how local livelihoods which relied on border crossing have had to
adapt to the extension of the Angolan state to the border. Finally, it shows how local
patterns of mobility in the borderlands are changing as the border is ‘normalised’.
Before launching into these findings, I should provide some more details on the
research process. The initial research was conducted over twelve months in 1996–97. The
study was originally framed around a research question about the repatriation of selfsettled Angolan refugees. However, from the outset my approach was to call into question
people’s motivations for moving into Angola, so my enquiry focused precisely on questions
of identity, livelihoods and mobility. Moreover, since this study rejected the assumption that
movement to peaceful Angola would only be a concern for refugees, I consistently asked the
same questions of both Angolans and Zambians in the border villages. The study used a
range of methods to investigate the nature of the cross-border linkages and movements,
people’s contacts with Angola, the integration of refugees and views towards repatriation.
The dataset included structured interviews with 195 people, both Zambian and Angolan, key
informant interviews, narratives of visits to Angola and observations in the field. Over the
next decade I maintained sporadic contact through correspondence, but my first
opportunity to return to Kanongesha only came in 2008, when I made a very short visit,
sufficient only for a very limited number of interviews. A third trip in 2010 over six weeks
gave me the chance systematically to find out the whereabouts of the original respondents
(gathering data on 96 per cent of the group), re-interview those still in Kangonesha (25 per
cent of the group) and observe more of the changes in the area over the last decade.
4.1 Borderland identifications
For Angolans moving into Zambia in the pre-colonial and colonial periods their journeys
were often one-way, and newcomers settled to become indistinguishable from those they
found on arrival. For those who came as refugees after Zambian independence, this avenue
of settlement was formally closed in the government’s eyes. Nonetheless, following the
historical precedents, many have been able to follow it with the full collaboration of other
villagers and the chiefs.
A large proportion of the people born in Angola who are now residing in Mwinilunga
entered Zambia as refugees. Under Zambian law, refugees must register with the
government authorities, who issue them with a refugee identity card and require them to
live within designated settlements, unless they have special leave to reside elsewhere.
Moreover, Zambian law provides no legal avenue for a person granted refugee status to
become a resident and in due course a citizen: once a refugee in Zambia, always a refugee.
Refugee status is passed on down the generations, so the child of refugees will also be seen
IMI Working Papers Series 2012, No. 63
11
as a refugee. This is the formal position of the Zambian state, which is restated in
government offices from Lusaka to the local district council.
Hence, the official policy has always been that refugees should register and move
away from the borderlands to the Meheba refugee settlement, which is near Solwezi.
However, any pressure that the government exerts on people to move has to be mediated
through the chiefs. In a number of cases, in particular in Kanongesha, the chiefs did not cooperate at all. For example, in 1990 government officials from Solwezi came to Kanongesha
and announced to the people that all refugees had to go to Meheba or they would be sent
back to Angola. This threat was immediately contradicted by the senior chief at the time,
who stressed that those who wanted to stay in his area would be welcome as his people.
Moreover, as the years have passed, many of those who have the formal status of
refugees have been able to acquire a national registration card as citizens, effectively
making it impossible for outside bureaucrats to distinguish Angolans from Zambians in the
borderlands (see Bakewell 2007). In particular, as refugee children have grown up, village
headmen, chiefs and local officials have colluded to vouch for them as ‘children of the
village’ who are entitled to Zambian identity papers. Even those who have not obtained
Zambian papers have reported that this causes no problems as long as they remain within
the chieftainship. It is only when they move beyond the borderland – for example travelling
on the main road to the Copperbelt – that they may be challenged to produce papers.
With the end of the war in Angola, the focus of the Zambian state and the
international community shifted towards the repatriation of Angolan refugees, including
those living in the border villages. UNHCR and the Zambian government attempted to
register self-settled refugees for return to Angola between 2003 and 2006. Teams of officials
visited the villages and asked for those who were interested in repatriating to Angola to
come forward. None of the respondents reported any pressure being exerted on people to
register from their neighbours, the headmen, the senior chief or even the officials
conducting the exercise. One Angolan couple said that they personally knew the officials
involved, invited them to their house during the registration process and took their advice
not to repatriate. The senior chief reminded people about the lack of health services and
education in Angola and discouraged people from registering. Therefore, it seems
reasonable to believe that those few who came forward did so voluntarily.
In these borderlands a rather robust convention of self-settlement of refugees has
emerged. Indeed UNHCR statistics suggest a much larger drop in the number of refugees
than can be explained by their figures for repatriation; this suggests that the self-settled
refugees may have been ‘written off’ the books.2 In a sense, the convention has outweighed
the social contract. Despite government policy which does not accept any integration of
refugees, there is evidence of very successful integration in Mwinilunga District. Since this
has come about through the exercise of borderland conventions, it remains
unacknowledged. However, this convention of informally accepting Angolans (refugees) will
inevitably be challenged now that the war has ended.
2
UNHCR assisted the return of 74,000 Angolans from Zambia between 2003 and 2007. According to UNHCR’s
statistics the number of Angolan refugees fell from the peak of 220,000 at the end of 2001, of which only
90,000 were receiving assistance, to 43,000 by the end of 2006, of which 18,500 were receiving assistance.
This suggests that the number of self-settled refugees has fallen by over 100,000.
12
IMI Working Papers Series 2012, No. 63
This effective ‘legal’ – in the sense of securing documents but technically illegal –
integration during the war did not result in people necessarily seeing themselves as
becoming Zambian. Some would make no distinction between being Zambian or Angolan.
Others referred to their nationality in pragmatic terms, for instance, saying: ‘while I am in
Zambia, I am a Zambian, but if I go to Angola, I will be Angolan’ (man 10.02 interviewed
25/1/97); ‘this time I am Zambian as my wealth is from Zambia [i.e. house, and fields], but I
will become Angolan again’ (man 12.00 interviewed 25/1/97); or, ‘I am a Zambian because I
stayed so long and have identity papers from here’ (man 5.06 interviewed 24/7/97).
For others, their nationality was a reflection of their history, a given fact of life. They
described themselves as Zambian or Angolan according to where they had been born or
brought up, or where their parents had come from. It was neither something to be adapted
to the circumstances nor revealing a sense of strong identification with their country.
Then there were those for whom a declaration of nationality expressed an emotional
bond to the country. They referred to feeling their nationality in the heart, or being ‘100 per
cent’, ‘full time’ or ‘pure’ Zambian or Angolan. In this sense, to say one is Zambian or
Angolan is not concerned so much with a legal status but a declaration of a person’s
identity in the ‘strong’ sense of a consistent presentation of oneself across time and in
different contexts (Brubaker and Cooper 2000: 10). Such people remained attached to
Angola despite living in Zambia for many years. The point here is that in 1997 during the
war, all these views of nationality were expressed freely and were perfectly acceptable.
There was no expectation that Angolans should renounce their Angolan-ness, even if they
acquired Zambian identity papers.
Has this convention of flexible citizenship that was prevalent in the borderlands been
sustained since the end of the war? The evidence from post-war fieldwork suggests that it
has not. As one man put it in 2008:
Now I say I am Zambian. If I stay in Zambia, I am Zambian. I have an NRC
[Zambian National Registration Card]. Before, I used to call myself Angolan as I
used to go there to hunt and fish. Now the people who went back are Angolan,
and those who stay here are Zambian. (Man 14.02 interviewed 9/3/2008)
Similarly, one woman who in 1997 had spoken enthusiastically of her plans to return to
Angola to stay should peace be established and schools reopened. At that time she
described her national identity as follows: ‘where I am staying that is my place, so at the
moment I am Zambian’. By 2008, she had abandoned any idea of going to Angola, saying of
her and her husband, ‘it is no longer in our minds to be Angolan’; in 2010 she said ‘I have
forgotten about Angola’ (woman 2.33 interviewed 19/6/1997, 9/3/2008 and 2/09/2010).
Her husband showed an even stronger shift in terms of his self-identification. In 1997,
he said that he was ‘full-time’ Angolan and planned to return as Angola ‘is my country’ (man
2.32 interviewed 26/02/1997). In 2010, these responses were forgotten:
I have been in Zambia 15–20 years, so I see myself as Zambian rather than
Angolan. When I came to Zambia in 1984, I said I would never go back to
Angola but stay here in Zambia, so I said I was Zambian (man 2.32 interviewed
3/09/2010).
All those who had described themselves as Angolan in 1997 and were re-interviewed in
2010 described themselves as Zambian; a couple of them denied ever having changed
IMI Working Papers Series 2012, No. 63
13
nationality. As participants in one village discussion said, ‘those who came from Angola have
become Zambian and have their NRCs [Zambian National Registration Cards]’ (village group
discussion 18/08/2010) and this was confirmed by individual respondents.
It seems then that the term ‘Angolan’ has shifted from being associated with refugees
and the stigma of war to referring to a ‘normal’ nation state. During the war, it was possible
for people in Zambia to declare themselves to be Angolan without further explanation for
their presence. Today, that has changed and those who remain in Zambia only refer to
themselves as Zambian. There now appears to be a greater correlation between people’s
declared nationality and the nationality which is officially recognised through their identity
papers.
4.2 Borderland livelihoods
Even after nearly forty years of war, the destruction of its infrastructure and the memories
of horrific violence which many people suffered there, Angola is often seen as a place of
wealth and opportunity by the people of the upper Zambezi. Their livelihoods are largely
based on subsistence cultivation, hunting, fishing and gathering, and the under-populated
bush in Angola has supplies of bush meat, fish, honey, mushrooms and caterpillars, beyond
anything available in Mwinilunga. During the war there were trading opportunities,
especially since Alto Zambeze district was cut off from the rest of Angola. In 1997, many
villagers in Kangongesha complained that the only way for them to obtain cash was to cross
into Angola.
This hunting and gathering was undertaken either in the deep bush close to the
border where people hoped to avoid any contact with UNITA or through negotiation with
the rebels. For example, hunting for bush meat was nearly impossible in Zambia as the only
animals remaining are in game management areas, a long way away and policed by Zambian
game guards who deal roughly with poachers. In Angola, there was no effective state and so
no formal control of hunting. Instead, hunters would come to deal with UNITA soldiers to
provide them with a portion of the kill. Some hunters also reported that they rented guns
from UNITA, while others used locally made muzzle loaders (with home-made ammunition),
or wire (made from bicycle brake cable) and nylon string (made from old maize sacks) to
make traps. It was not uncommon for these hunting parties to go deep into the bush for
days and still return with very little, but a successful trip could be very profitable. Likewise,
many people crossed from Zambia into Angola to reach the larger rivers and the better
supply of big fish. On both hunting and fishing trips, encountering UNITA could easily result
in people losing their whole catch (and all their other belongings) to the rebels.
The thicker bush in Angola also has an abundance of other natural resources that are
coming under more pressure in Zambia. Many beekeepers set hives over the border and
honey hunters were more likely to find natural hives in Angola. The women who collected
caterpillars and mushrooms also went across the border where there were more to be
found. All these activities were generally limited to the bush close to the border, and tales of
such trips did not usually involve any contact with UNITA or large settlements; nonetheless,
the fear of encountering the rebels was always in their minds.
In 1997, there was also continuous cross-border trade. The ‘bush’ products of meat,
fish, honey, caterpillars and mushrooms could all be profitably sold in the Zambian
Copperbelt. While many people from Kanongesha would cross into Angola to gather these
14
IMI Working Papers Series 2012, No. 63
bush products, it was a risky and time-consuming business and there are many who
preferred to leave the gathering to people in Angola and to go across only to buy the goods
from them. During the war, this trade was very important for people living on the Angolan
side of the border as they were almost completely cut off from the rest of Angola, especially
since the bridge across the Zambezi had been destroyed in the 1980s. It was difficult for
local people to move out of UNITA-controlled Alto Zambeze into areas controlled by the
Luanda government. As a result, most of the supplies for people in the area came from
Zambia. Within this borderland, much of the trade was conducted through a barter system,
and people usually returned with meat and fish. The traders from Kanongesha used back
routes where they had no contact with the Zambian authorities and paid no duties. If they
went to markets in the towns of Angola, they were likely to have to pay ‘taxes’ to UNITA.
With the end of the war, it might have been expected for these exchanges with Angola
to expand as it became safer to cross. However, the evidence from those living in
Kanongesha suggests that the opposite is true. The use of the ‘private roads’ has declined
and people now need to obtain papers and are more likely to cross via the ‘immigration
routes’. In 2008, people reported that it was now much tougher to negotiate with the
border patrols and immigration officials that they now find in Angola. As one respondent
put it, it is much harder to go hunting and fish now ‘the government is at work’. By 2010,
the Angolan government had installed a unit of its border guards (Policia Frontera, referred
to as the PF) right next to the boundary on one of the main back routes into Angola. These
border guards often came across into Zambia to obtain basic supplies, drink and socialise. In
Zambia, they were received in a friendly fashion, but they were widely seen in a different
light when they were in Angola on patrol; there, they were a threat and many people
complained about their seizing goods and harassing people who crossed without papers or
were caught hunting.
It was acknowledged that although encounters with the PF might be expensive, they
were not lethal. This is in stark contrast to the days of UNITA, who were known to kill people
on occasion. Nonetheless, it is clear that the people of Kanongesha had developed a
working relationship with UNITA over the years (despite the fact that many of them had fled
Angola to escape when UNITA captured Alto Zambeze in the early 1980s). Many felt they
could negotiate with them and do business, whether it was hiring guns for hunting or paying
them ‘taxes’:
UNITA was good – FAPLA [Forças Armadas Populares de Libertação de Angola] is
difficult and does not understand the people. (woman 70+ interview 22.03
18/08/10)
During the war there was lots of trade, but it seems more difficult now. UNITA
made friendship with people, but people have not made friends with the soldiers
there now. (man 42 interview 26.00 18/08/10)
Now it is hard to cross as the laws are binding. During UNITA time it was easy as
UNITAs were also looking for foods so they could not refuse people. (woman 40
interview 14.04 3/09/10)
People are still struggling to build up a similar flexible and accommodating relationship with
the Angolan government officials, who are much more recent arrivals in the area. However,
the cosy contacts may never be replicated with the government. UNITA needed people to
IMI Working Papers Series 2012, No. 63
15
bring supplies from Zambia for their survival in the area; the PF can come into Zambia to do
their business at will and they also receive government supplies by helicopter.
Looking at it from the Zambian perspective, the end of the war may have brought little
change to the cross-border exchanges. It may be seen as an everyday tale of border folk
using informal channels for smuggling, which has changed little over the last fifteen years.
However, from the Angolan point of view, there has been a much more fundamental shift.
During the war, the social contract, enshrined in formal law and the constitution, was
effectively suspended in this remote corner of UNITA’s territory. Instead, a reasonably
stable set of conventions emerged which shaped people’s expectations of encounters with
the UNITA military and civilian authorities, making it possible for them cross and do
business. Far from working against the quasi-state established by UNITA, these conventions
helped to sustain it.
The shift from peace to war has effectively undermined these conventions and
reinstated the social contract on the Angolan side. The extent to which the villagers of
Kanongesha and other border areas can establish new conventions which enable them to
continue their cross-border livelihoods may be very limited. As a result, the extensive
borderland in which they could secure their livelihoods that reached as far as Cazombo
during the war may have irrevocably shrunk to the narrow margins alongside the streams
marking the border itself.
4.3 Borderland movements
The strong historical, cultural, social and economic links between Angola and Zambia were
reflected in the steady flow of the people of Kanongesha back and forth across the border.
During the war, these movements ebbed and flowed according to the situation in Angola. In
periods of relative calm more people were visiting Angola; about 58 per cent of men and 22
per cent of women interviewed in 1997 had visited Angola in the past year for one reason or
another. Many of these were short trips, often confined to the bush, avoiding any contact
with the Zambian authorities or UNITA.
Some people were adamant that they were only interested in visiting Angola and they
would never settle there, whatever the situation in the future. However, many people
(about 30 per cent of those interviewed) expressed great interest in settling in Angola if it
achieves a stable peace. Across the borderlands (and elsewhere in Zambia) the draw of
Angola was widely recognised. Villagers and officials were anticipating a significant drop in
population in Kanongesha and the other border areas of Zambia when the situation
improved in Angola.
Before the end of the war, some had already begun the long, slow, flexible process of
moving to settle in Angola. This started with going back and forth continuously in ever
longer visits, and in due course establishing new fields and even a new household across the
border. Hence, some villagers were described as having ‘one foot in Zambia and the other in
Angola’, and their weight shifted from foot to foot all the time. This was not the one-off
event of ‘return’ migration envisaged in the discourse of refugee repatriation. Rather it was
a process with no clear beginning or end.
Since the end of the war, there has certainly been a shift of population to Angola and
that shift seems likely to continue as conditions there improve. However, the movement is
16
IMI Working Papers Series 2012, No. 63
not on the scale of the depopulation of the border areas which was anticipated by many in
1997. Half of those who had planned to move to Angola – these included both people from
Angola and Zambia – were still living in Zambian Kanongesha in 2010, and those interviewed
seemed much more reticent about going there:
The people who were bringing fish and meat from Angola have gone back there.
Those who stayed have put their mind to farming rather than catching fish and
meat. (man 58 interview 2.32 3/09/10)
Since the war ended, people from Angola in Zambia feel comfortable and have
started developing. The future is in Zambia not in Angola. (man age 60+
interview 1.02 15/08/10)
The increased difficulty in crossing the border, especially the perceived harassment from
Angolan border patrols, and the lack of basic services, in particular schools, were among the
reasons people cited for their change of plans. Some of those who had attempted the move
then found that it did not work out as expected, so they had returned to Zambia. Even
interest in visiting Angola appears to have declined. Eighteen of those re-interviewed had
expressed an interest in visiting Angola in 1997, but only half of these still had any ideas of
visiting Angola when asked in 2010.
As a result, the level of cross-border movement appeared to have declined since the
war in Angola. Again, the conventions which governed movement across the borderland –
continuously reproduced in moments of interaction between villagers, Zambian officials or
UNITA – appear to have broken down with the extension of the Angolan state to the border
with Zambia. The informal opportunities for migration appear to be giving way to
programmes of repatriation orchestrated by the UNHCR and the Angolan and Zambian
governments, and ‘regular’ processes of immigration requiring official permits and
payments.
5 Conclusion
Here it has been possible only to sketch some of the possible transformations in people’s
relationship with the border as Angola consolidates its peace. Certainly, my (perhaps naïve)
expectations that the end of the war would encourage more cross-border trade and
movement have been confounded. Instead, the fundamental changes in the political and
security situation in Angola have rippled out to the borderlands where they have unsettled
the long-standing practices of informal but patterned negotiation and accommodation that
operated during the many years of war.
The border infrastructure of immigration and customs posts, paperwork and other
controls is increasing. This facilitates the movement of those who move within the
framework of the social contracts that are recognised by the governments on both sides. In
this sense the border can be seen as more open. Moreover, it may be taken as a positive
sign of development (c.f. Miles 2005) which allows people to put aside their precarious
cross-border livelihoods in favour of new opportunities. However, in another sense, the shift
from war to peace in Angola is closing down the borderlands for villagers living in
Kanongesha, who are far from the formal crossing points and are largely excluded from the
formal border regimes.
IMI Working Papers Series 2012, No. 63
17
The onset of peace also appears to be increasing the significance of national identity
on each side of the border, bringing to an end the ease (or at least informality) of movement
that was possible in pre-colonial times and was able to continue as long as Alto Zambeze
remained largely a no-man’s land beyond the reach of formal state structures.
This process of simultaneous open and closing of the borderlands along with the
sharpening of the lines defining identity and movement may be a reflection of the peculiar
circumstances of this Zambia–Angola border. Nonetheless, it does resonate with the
analysis of other cases found in the literature. To some extent, the separation between the
formal law and policy and actual practice that it reveals is well-trodden ground. However, I
argue that the analysis presented here brings to light a more profound link between these
conventions and the core functioning of the state than is often recognised.
Before pursuing this discussion, it is important to stress that the border has not been a
product of war. The Angolan civil war has been a struggle within the given borders of the
country – the only claim for secession has been in Cabinda, the small oil-rich enclave far
away on the Atlantic coast. The Angola–Zambia border may have been breached at various
times during the forty years of conflict, with occasional incursions, first by the Portuguese
and then UNITA and FAPLA, but always primarily in pursuit of opponents. It has never been
a dispute over territory with Zambia. Moreover, on the Zambian side of the border, the
state is neither the aggressor nor is it using violence to displace or determine insiders and
outsiders. Even on the Angolan side, during most of the period of the study, the people
crossing the border have been of rather marginal interest to the authorities, whether the
UNITA ‘quasi-state’ or, since the end of the war, the Luanda government. In this region,
violence has not featured as a dominant means of state making. Of course violence lay at
the centre of the struggles for the Angolan state, but the borderlands were for the most
part on the periphery of the conflict; hence, they could be used as a place of refuge.
It might be tempting to see the conventions described here simply as a reformulation
of the concept of customary or traditional law, which may run in parallel with the statutory
law.
Customary law is a body of rules governing personal status, communal resources, and
local organization in many parts of Africa. It has been defined by various ethnic groups for
their internal organization and administration. Customary law is recognized by the courts
and exists as a second body of law (in addition to statutory law) governing citizens in the
countries of Sub-Saharan Africa (Joireman 2008: 1235).
In Zambia, this is embedded in the constitution. The chiefs have jurisdiction to settle
many minor cases, such as theft, divorce and so forth in the local courts. In Angola, there is
no such formal recognition of the chiefs’ positions but they still wield authority to settle
some local disputes. In neither country can the conventions around citizenship, cross-border
trade and migration that are discussed here be seen as part of customary law; they lie far
beyond its jurisdiction.
To a large extent the conventions described in the chapter here have become
institutionalised; do they reflect more than the oft-described discrepancy between law and
institutionalised practice? For example, Hughes (1999) appears to describe a similar process
of local integration of Mozambican refugees in Zimbabwe in the 1990s, which ran counter to
Zimbabwean law. He shows how the local chiefs encouraged Mozambicans to settle on
areas of their traditional land which had been incorporated within the national park and
18
IMI Working Papers Series 2012, No. 63
were the subject of ongoing disputes with the state. Not only did this mean that the
Mozambicans were in a marginal position, but they also served an important local political
purpose. The settlement of ‘refugees’ was then seen as a way of pursuing a contest for
power and resources with the Zimbabwean state.
There is little evidence of such a process in this part of the Zambia–Angola border.
Further south, in the district of Chavuma, there has been more suggestion that the
settlement of refugees has played into a long-standing land dispute (see Hansen 1979).
However, there the dispute is between Lunda and Luvale groups, a local contest between
senior chiefs, each appealing to the state to support their case. The settlement of refugees
is not being used to make a claim against the state. That said, the presence of Angolans, in
as far as it represents an increase in population, is used by the chiefs to increase their
prestige and make claims for government resources, in particular services such as schools,
clinics and roads.
The process of incorporation by the Zimbabwean chiefs might be seen as an example
of a ‘twilight institution’ as discussed by Lund (2006a; 2006b). However, the Zambia–Angola
borderland conventions that I describe do not correspond so neatly. They are not the same
as local institutions such as chieftainships, vigilante groups, local elite groups, that claim
authority – often backed by historical narratives and territorial markers – to act in particular
localities (Lund 2006b) in opposition to the state, nor do they claim separation from the
state. Indeed if we take Lund’s conception of the state being ‘formed as a combination of
people’s everyday encounters with representatives of the state and its representations’
(Lund 2006b: 689), it is reasonable to suggest that the conventions described here reflect
the operation of state in the local area. They effectively support and reproduce the state at
a most fundamental level. They are not simply concerned with local politics, but these are
local conventions that tie in directly to central issues of the state – in particular, who can
come and go, and who is accepted as a citizen. This is the stuff of ‘government institutions
that claim to be the embodiment of the state’ (Lund 2006b: 699).
Here we are concerned with public authority embedded within the state (or quasistate in the case of UNITA) – not reflecting how things work in the ‘face of obvious state
failure and impending collapse’ (Lund 2006a: 674) but how local interactions interpret and
enable the state to function. There is little evidence of people challenging the legitimacy of
the state to exert its power. The question is how can it do work with such limited resources
and weak manifestations. The actors who play out these conventions are agents of the
state, often acting in their official roles, but making the best of what opportunities they
have, both to fulfil their tasks given the resources at their disposal and also for their
enrichment and self-aggrandizement. Their acceptance of cross-border movement and
settlement is not a form of resistance to the state but the form of governmentality in the
borderlands. Unlike twilight institutions, these conventions are not competing with the
social contract (Lund 2006b: 699); they are interpreting it.
As I have shown in this chapter, the end of the war in Angola has unsettled the
conventions shaping people’s identity, livelihoods and mobility in the Zambia–Angola
borderlands. Observing the changes in people’s expectations and practices casts a light on
the profound role these local conventions play in the art of state-making in a remote corner
of Africa. While the particular conditions which brought these interactions to light might be
unique to this case, we should expect to find similar conventions that effectively mould
citizenship and border crossing emerging in many other borderlands.
IMI Working Papers Series 2012, No. 63
19
References
Anderson, James and Liam O'Dowd. 1999. "Borders, Border Regions and Territoriality:
Contradictory Meanings, Changing Significance." Regional Studies 33:593 - 604.
Bakewell, Oliver. 2000. "Repatriation and Self-settled Refugees in Zambia: Bringing Solutions
to the Wrong Problems." Journal of Refugee Studies 13:356-373.
—. 2007. "The Meaning and Use of Identity Papers: handheld and heartfelt nationality in the
borderlands of North-West Zambia." International Migration Institute, Oxford.
Baud, M. Michiel and W. Willem Van Schendel. 1997. "Toward a Comparative History of
Borderlands." Journal of World History 8:211-242.
Bender, G.J. 1978. Angola Under the Portuguese: the Myth and the Reality. London:
Heinemann.
Brubaker, Rogers and Frederick Cooper. 2000. "Beyond "identity"." Theory and Society 29:147.
Brunet-Jailly, Emmanuel. 2005. "Theorizing Borders: An Interdisciplinary Perspective."
Geopolitics 10:633 - 649.
Cheke Cultural Writers. 1994. The History and Cultural Life of the Mbunda Speaking Peoples.
Lusaka: University of Zambia.
Clapham, Christopher, Jeffrey Herbst, and Greg Mills. 2006. "Big African States: Angola,
Sudan, DRC, Ethiopia, Nigeria, South Africa." Pp. 301. Johannesburg: Wits University
Press.
Clarence-Smith, W.G. 1979. Slaves, Peasants and Capitalists in Southern Angola 1840-1926.
UK: Cambridge University Press.
Das, Veena and Deborah Poole. 2004. "Anthropology in the margins of the state." Pp. ix, 330
in School of American Research advanced seminar series. Santa Fe; Oxford: SAR Press;
James Currey.
Donnan, Hastings and Thomas M. Wilson. 1999. Borders: frontiers of identity, nation and
state. Oxford: Berg.
Flynn, Donna K. 1997. ""We are the border": identity, exchange, and the state along the
Bénin-Nigeria border." American Ethnologist 24:311-330.
Hansen, A. 1979. "Once the Running Stops: Assimilation of Angolan refugees into Zambian
border villages." Disasters 3:369-374.
Henderson, L.W. 1978. Angola: Five Centuries of Conflict. Ithaca and London: Cornell
University Press.
Hughes, David McDermott. 1999. "Refugees and Squatters: Immigration and the Politics of
Territory on the Zimbabwe-Mozambique Border." Journal of Southern African Studies
25:533-552.
International Crisis Group. 2003. "Angola: Exorcising Savimbi's Ghost." Current History
102:206-214.
20
IMI Working Papers Series 2012, No. 63
Joireman, S. F. 2008. "The Mystery of Capital Formation in Sub-Saharan Africa: Women,
Property Rights and Customary Law." World Development 36:1233-1246.
Kolossov, Vladimir. 2005. "Border Studies: Changing Perspectives and Theoretical
Approaches." Geopolitics 10:606 - 632.
Lund, Christian. 2006a. "Twilight institutions: An introduction." Development and Change
37:673-684.
—. 2006b. "Twilight Institutions: Public Authority and Local Politics in Africa." Development
and Change 37:685-705.
Megoran, N. 2006. "For ethnography in political geography: Experiencing and re-imagining
Ferghana Valley boundary closures." Political Geography 25:622-640.
Miles, W. F. S. 2005. "Development, not division: local versus external perceptions of the
Niger-Nigeria boundary." Journal of Modern African Studies 43:297-320.
Newman, D. 2006. "The lines that continue to separate us: borders in our 'borderless'
world." Progress in Human Geography 30:143-161.
Nugent, Paul. 2002. Smugglers, secessionists and loyal citizens on the Ghana-Togo frontier:
the life of the borderlands since 1914. Athens; Oxford; Legon: Ohio University Press;
James Currey; Sub-Saharan Publishers.
—. 2010a. "Sizing up Asymmetry: State Logics and Power Dynamics in the Senegambian and
Ghana-Togo Borderlands." in Borderlands Theory in Africa - ABORNE Annual Conference.
University of Witwatersrand, South Africa: African Borderlands Research Network.
—. 2010b. "States and Social Contracts in Africa." New Left Review 63:35-68.
Pritchett, James Anthony. 2001. The Lunda-Ndembu: style, change, and social
transformation in South Central Africa. Madison: University of Wisconsin Press.
Pycroft, C. 1994. "Angola - 'the Forgotten Tragedy'." Journal of Southern African Studies
20:241-262.
Raeymaekers, Timothy. 2009. "The silent encroachment of the frontier: A politics of
transborder trade in the Semliki Valley (Congo-Uganda)." Political Geography 28:55-65.
Roberts, A. 1976. A History of Zambia. NY: Africana Publishing Company.
Sangambo and M.K. 1984. The History of the Luvale People and their Chieftainship. Zambezi,
Zambia: Mize Palace.
Turner, V. 1957. Schism and Continuity in an African Society: a study of Ndembu village life.
Manchester: Manchester University Press.
von Oppen, Achim. 1995. Terms of Trade and Terms of Trust: the history and contexts of
pre-colonial market production around the Upper Zambezi and Kasai. Hamburg: Lit.
Zartman, I William. 2010. "Understanding Life in the Borderlands: Boundaries in Depth and
in Motion." Athens, Georgia: University of Georgia Press.
Zeller, Wolfgang. 2009. "Danger and Opportunity in Katima Mulilo: A Namibian Border
Boomtown at Transnational Crossroads*." Journal of Southern African Studies 35:133 154.
IMI Working Papers Series 2012, No. 63
21